DESCRIPTION (provided by investigator): Despite increasing concern about the growing problem of abuse/dependence due to misuse of analgesic medications, little is known about the nature and extent of this public health problem in the United States. Extramedical analgesic initiation has been on the rise during the past decade, but the characteristics of analgesic users as well as the potential pathways associated with extramedical analgesic use and abuse/dependence are still understudied. The objectives of this study are to : 1) test for changes in characteristics of extramedical analgesic users over time, 2) test for potential causal pathways between psychiatric disorders and extramedical analgesic use and abuse/dependence, 3) estimate the distribution of symptoms of extramedical analgesic abuse/dependence in the U.S. household population, 4) test for categories of problem extramedical analgesic use based on observed clustering of symptoms of analgesic abuse/dependence and identify factors associated with analgesic abuse/dependence. We use data from two major surveys of drug use and related problems in the U.S: the National Survey on Drug Use and Health (NSDUH), and the National Epidemiologic Survey on Alcoholism and Related Conditions (NESARC). After initial exploratory data analyses, we will use logistic regression to estimate associations of extramedical analgesic use with demographic characteristics, psychiatric comorbidity and patterns of other drug use and to test whether these associations change over time. We will estimate the prevalence of extramedical analgesic abuse/dependence over time and test whether the conditional probability of abuse/dependence among extramedical analgesic users has changed over time. We will use Cox proportional hazards models with time- dependent covariates to test for potential causal pathways between psychiatric disorders and extramedical analgesic use, abuse and dependence. In addition, we will estimate incidence and prevalence of disorders and individual symptoms overall in the population and use generalized estimating equations to compare profiles of symptoms of abuse/dependence across subgroups of extramedical analgesic users and between current and past extramedical analgesic users. We will use latent class analysis to test for the best fitting latent class model using the symptoms of analgesic abuse/dependence and compare these classes to existing diagnostic categories, and we will test for variation in latent class structure by subgroups of drug use - e.g., compare latent class structure for analgesic users without other opiate use versus those with heroin use. Latent class regression will be used to investigate how class membership varies in relation to characteristics of interest (e.g., demographic characteristics). Finally, we believe that our study will bring significant public health contributions to help develop prevention and treatment strategies in order to decrease extramedical analgesic use, abuse and dependence. The proposed research will shed light in the understanding of potential problems related to extramedical analgesic use and of causal pathways between psychiatric disorders and extramedical analgesic use. It will contribute to the development of prevention, intervention and treatment for problem extramedical analgesic users. [unreadable] [unreadable] [unreadable] [unreadable]